Scheduling best-effort and real-time pipelined applications on time-shared clusters

Two important emerging trends are influencing the design, implementation and deployment of high performance parallel systems. The first is on the architectural end, where both economic and technological factors are compelling the use of off-the-shelf computing elements (workstations/PCs and networks) to put together high performance systems called clusters. The second is from the user community that is finding an increasing number of applications to benefit from such high performance systems. Apart from the scientific applications that have traditionally needed supercomputing power, a large number of graphics, visualization, database, web service and e-commerce applications have started using clusters because of their high processing and storage requirements. These applications have diverse characteristics and can place different Quality-of-Service (QoS) requirements on the underlying system (low response time, high throughput, high I/O demands, guaranteed response/throughput etc.). Further, clusters running such applications need to cater to potentially a large number of users (or other applications) in a time-shared manner. The underlying system needs to accommodate the requirements of each application, while ensuring that they do not interfere with each other.

This paper focuses on the CPU resources of a cluster and investigates scheduling mechanisms to meet the responsiveness, throughput and guaranteed service requirements of different applications. Specifically, we propose and evaluate three different scheduling mechanisms. These mechanisms have been drawn from traditional solutions on parallel systems (gang scheduling and dynamic coscheduling), and have been extended to accommodate the new criteria under consideration. These mechanisms have been investigated using detailed simulation and workload models to show their pros and cons for different performance metrics.